New results boost NuQ for lung and CRC

VolitionRx has developed the blood-based cancer diagnostic test, NuQ, which detects the structural changes to cell-free nucleosomes in the blood stream that occur in response to cancer or a precancerous lesion. Nucleosomes are a combination of DNA and protein that protect DNA in the cell, as well as determining how tightly it is packed. A number of chemical changes occur to nucleosomes to control the degree of packing and, by extension, the expression of the associated genes. During some of the changes that occur during the development of cancer, these nucleosomes become modified to allow for the expression of genes that lead to runaway cancer growth, and these modifications can be detected by NuQ tests. A typical screen consists of a panel of multiple independent NuQ tests, each testing for a different nucleosome modification, the combined results of which give predictive power. Additionally, these results may be adjusted for the age and/or gender of the individual to improve the accuracy of disease prediction.

An advantage of NuQ is that it can be performed using a simple blood test and because it uses routine ELISA-based assays, it can be integrated into pre-existing blood testing panels at low cost. Initial efficacy of the test was demonstrated using databases of blood samples from CRC and control patients. However, recent clinical efforts have been focused on demonstrating this efficacy, with real patients presenting at the hospital with symptoms. Additionally, because the modifications to nucleosomes detected by NuQ occur in numerous cancers, there has been an effort to test its capacity to detect other malignancies. To this end, the company has detected CRC, non-small cell lung cancer (NSCLC) and pancreatic cancer in real patients with approximately 90% sensitivity and specificity across the board, with planned investigations in a wide array of other malignancies currently underway.

The company released data on a recent trial of 121 patients that were high risk or presented with symptoms of CRC at CHU Dinant Godinne – UCL Namur in Belgium. Two different testing panels were used: one testing for four different nucleosome modifications, but including an adjustment of the age of the patient; and one testing for five different modifications but without adjustments for age. Of these two panels, the four-test panel with age adjustments outperformed the five-test panel in early-stage cancer and for all stages combined.

23 patients in the study were confirmed to have CRC, of which 21 were correctly identified with the age-adjusted NuQ panel for a 91% sensitivity rate. There were two false positives for a specificity of 90%. Both of these readouts are superior to the data gathered from previous retrospective studies with 81-84% sensitivity and 76-78% specificity. There are multiple different explanations for the increase in test accuracy in the recent study. It is possible that blood samples freshly collected from patients were of higher quality and yielded stronger analytical results, but the potential for selection bias on the part of investigators cannot be ruled out. Patients in this study were preselected as individuals at high risk for CRC or presenting symptoms of CRC, and although all stages of disease were present in the study population, the distribution of CRC by stage was weighted towards later stages compared to other clinical studies. There is therefore a possibility that the increase in the accuracy of the test may result from a correlation between the physician’s assessment and the patients who perform best in the test.

Despite this, the age-adjusted NuQ panel performed similarly for all stages of disease. Other diagnostics such as the fecal immunochemical test (FIT) perform better for late-stage disease and there is some evidence, albeit with a limited number of patients, that Exact Sciences’ Cologuard performs best in early-stage disease. The age-adjusted NuQ panel also identified 67% of precancerous adenomas (polyps), consistent with previous studies, which significantly outperforms both the FIT test (24%) and Cologuard (42%).

The increase in sensitivity demonstrated in this study may improve the path to FDA approval, because it provides substantial differentiation from the FIT test. This issue was recently highlighted in a letter from the FDA to Epigenomics rejecting the PMA for the Epi proColon blood test. Originally in June 2014 the FDA requested compliance data comparing Epi proColon to the FIT test when rejecting its first PMA. Epi proColon only demonstrated 72% sensitivity, which is approximately the same as the FIT test, and specificity was significantly less at 81%. The FDA therefore asked the company to prove the value proposition for Epi proColon contingent on it being less objectionable than a fecal test consequently showing better compliance. The resulting compliance trial completed in August showed almost 100% compliance for Epi ProColon, but interestingly 88% compliance for the FIT test, which is much higher than historical rates. In November, 2015, the FDA reiterated its stance in a response to these data, rejecting the company’s PMA and again requesting additional information on compliance from Epigenomics.

The recent data from VolitionRx differentiate it significantly from the FIT test and there is therefore a reason to believe that it will be able to support a value proposition independent of compliance. These data are more similar to those submitted for approval of Cologuard, which did not require compliance data. However, the recent US Preventative Services Task Force (USPSTF) draft guidance regarding Cologuard highlighted the fact that it is still a fecal test like FIT, with the implied compliance limitations, and this was part of the reasoning for listing it as an alternative therapy. So, it might be in VolitionRx’s best interests to perform compliance studies to enhance its value proposition in the future, thinking past approval and towards inclusion into guidelines and payor acceptance.

VolitionRx has an ongoing trial investigating the NuQ test for detection of lung cancer. The company recently released interim results of the first 73 of the planned 240 patients referred to Liège University Hospital in Belgium for potential pulmonary disease. The single-arm study using a panel of four NuQ blood assays adjusted for smoking history identified 27 of 29 patients with NSCLC, with two false positives (out of 22 healthy patients) for a sensitivity of 93% and specificity of 91%.

The recent results represent a marked increase in accuracy compared to a pilot study previously completed in late 2014 using a panel of two NuQ tests on both sputum and blood. Sputum testing proved more effective in this study, identifying 18 out of 21 lung cancers (85%) compared to 16 (76%) identified via blood. The increase in the specificity of the blood test is likely due to the increase of NuQ tests (from two to four) in the panel. The success of the test in blood is significant because of the increase in convenience to the patient and the doctor associated with a blood draw as opposed to sputum testing, which may require an invasive bronchoscopy to obtain a sample.

The specificity of the blood test and its capacity to identify cancer in a background of other lung diseases is critical to its commercial success, because a significant contributing factor to the morbidity of lung cancer is that it is typically detected late in the progression of the disease. The detection of lung cancer from blood may allow for the routine testing of at-risk individuals, as there is currently a lack of effective non-invasive or low-risk detection methods. The consensus target population for lung cancer screening is people aged 55 to 76 (on average) with 30 pack years of smoking. This represents a potential market in the range of nine million individuals in the US.1

Historically, diagnosis of lung cancer has involved a combination of imaging and cell samples. Low-dose computed tomography (LDCT) has the highest success rate of imaging techniques for identifying cancer, but requires radiation exposure. The current NuQ lung results are very close to the sensitivity and specificity of LDCT, posing the possibility of a safer and more convenient alternative.

An alternative to LDCT is pre-screening with a chest X-ray followed up with biopsy of suspicious lesions. This procedure has one of the highest chances of detecting lung cancer, but there is significant risk associated with lung biopsy, so many doctors perform a bronchoscopy to obtain a sputum sample for cytology. However, retrospective studies have identified that sputum cytology misses approximately one-third of cancers (66% sensitivity).2

Veracyte has developed a lung cancer diagnostic test to address this issue, Percepta, which was CLIA waived in early 2015. The test uses gene expression analysis from cells extracted during a bronchoscopy to identify cancer, and has exceptionally high sensitivity (97%), but low specificity (47%). As such, it is targeted at the 100,000 patients per year with inconclusive sputum cytology, which positions it for a different niche from NuQ.

Epigenomics has developed Epi proLung BL Reflex, a test that measures DNA methylation in bronchial aspirate to detect lung cancer. A blood-based test has also been in development, but showed very low sensitivity (62%). The recent NuQ lung cancer data significantly outperform the sensitivity of the Epi proLung (81%). Additionally, Epi proLung is based on inherently more complex genetic technology, which is more difficult and costly to implement than the already entrenched ELISA technology that is the basis for NuQ.

VolitionRx has had a strong track record engaging institutions and initiating collaborative studies. These efforts have resulted in a substantial number of ongoing clinical trials, the majority of which will have readouts in 2016. CRC remains a significant focus of the company, with three trials ongoing in collaboration with Hvidovre Hospital in Denmark, all of which have expected readouts in H116. Among these is the first study of NuQ’s unique capacity to detect colorectal adenomas, the precancerous polyps that significantly increase the risk of developing CRC. In previous trials examining CRC, NuQ detected 67% of adenomas, which is significantly better than existing non-invasive tests. The upcoming data on adenoma may be used by the company to optimize the NuQ panel to further improve the detection of these lesions. Data are expected in this trial in H116.

Additionally, the company continues to expand and support the development of NuQ testing panels for other oncology indications. There are additional results from a second, confirmatory 600-person lung cancer study expected in H216, as well as expansions into prostate cancer, endometriosis and ovarian cancer. Early studies into CRC had the interesting, unintended consequence of identifying patients with other forms of cancer, otherwise unrelated to CRC. Currently, the exact extent of different cancers that can be detected by NuQ tests and the degree to which these test results can differentiate cancer types is unknown. To answer these questions, the company is gathering data from 4,700 people across a range of 27 different malignancies in a study at Bonn University, which should provide insight into the utility of NuQ tests for other cancer types. Data are expected for this trial in H216 and should be a significant catalyst for the company.

We have adjusted our valuation of the company from $189m or $10.48 per share to $197m or $10.90 per basic share, primarily due to expectations of greater expense control, which is mitigated by a lower cash balance. VolitionRx had $6.9m in cash at the end of Q315 and is likely to raise capital in early 2016.

VolitionRx had $6.9m in cash and cash equivalents at the end of Q315. With a burn rate of ~$2m per quarter, the company will need to raise capital in 2016. Sales and marketing expenses are expected to increase as the company invests in European commercial operations following CE mark approval of NuQ for CRC. R&D expenses are also expected to increase as the company initiates trials with the aim of US approval. We currently forecast $9.4m in R&D in 2016 and $10.3m in 2017, which is lower than our previous forecast as the company has continued to control expenses. If VolitionRx decides to go down the self-marketing route, we calculate that it would need to raise an additional $120m before profitability in 2021. Near term, we expect it to raise $25m in 2016 and $25m in 2017 (nominally attributed to debt as per our policy).